Revealing and incorporating implicit communities to improve recommender systems

Social connections often have a significant influence on personal decision making. Researchers have proposed novel recommender systems that take advantage of social relationship information to improve recommendations. These systems, while promising, are often hindered in practice. Existing social networks such as Facebook are not designed for recommendations and thus contain many irrelevant relationships. Many recommendation platforms such as Amazon often do not permit users to establish explicit social relationships. And direct integration of social and commercial systems raises privacy concerns. In this paper we address these issues by focusing on the extraction of implicit and relevant relationships among users based upon the patterns of their existing interactions. Our work is grounded in the context of item recommendations on Amazon. We investigate whether users' reply patterns can be used to identify these meaningful relationships and show that different degrees of relationships do exist. We develop global measures of relationship strength and observe that users tend to form strong connections when they are evaluating subjective items such as books and movies. We then design a probabilistic mechanism to distinguish meaningful connections from connections formed by chance and extract implicit communities. We finally show that these communities can be used for hybrid recommender systems that improve recommendations over existing collaborative filtering approaches.